CountBot is a lightweight, open-source AI Agent framework and hub tailored for Chinese users, enabling local deployment and long-term operation. It connects Large Language Models (LLMs), IM channels, workflows, and external tools, empowering ordinary users to interact with AI via natural language for task execution and automation.

How It Works

CountBot integrates roles, teams, workflows, memory, LLMs, and IM channels into a cohesive system. Its core architecture features an Agent Loop for ReAct reasoning and tool invocation, a multi-robot channel matrix for diverse IM integrations (WeChat, Feishu, Telegram, etc.), and layered configuration. It supports complex Agent Teams with pipeline, graph, or council collaboration modes, and connects to external execution tools like Claude Code. The system emphasizes local control, security, and long-term operation.

Quick Start & Requirements

Installation involves git clone, pip install -r requirements.txt, and python start_app.py. Basic requirements include Python and Git. Desktop builds are available for Windows, macOS, and Linux. Documentation is accessible at https://654321.ai/docs/.

Highlighted Details

• Extensive multi-channel support: Integrates with WeChat (ClawBot), Weibo, Feishu, DingTalk, QQ, Telegram, Enterprise WeChat, and XiaoZhi AI.
• External tool integration: Incorporates industry tools like Claude Code, Codex, and OpenCode into agent execution.
• Agent Team orchestration: Enables multi-role collaboration using pipeline, graph, or council modes for complex task decomposition.
• SkillsHub and IMA integration: Connects to Tencent Cloud SkillsHub for natural language skill management and IMA for knowledge base/note management.
• Model thinking control: A switch adjusts model "thinking intensity" for faster responses.

Maintenance & Community

Actively maintained with regular releases (latest v0.7.0) focusing on stability and features. Community interaction via QQ group (1028356423) and GitHub Issues.

Licensing & Compatibility

Released under the MIT License, permitting commercial use and integration into closed-source projects.

Limitations & Caveats

Specific hardware requirements for advanced features are not detailed. The focus on Chinese user habits may affect non-Chinese speaker UX. The project is under continuous development, indicating ongoing refinement.